Variety and diversity. Considerations fro empirical reserach on innovation

This paper explores the conceptualization of variety in the more recent evolutionary theories of innovation and technological change, and goes on to investigate the difficulties of meaningfully applying such a concept in empirical analyses of inter-firm differences. The main point relates to the interpretation of variety in technology and behaviors and how it can be ascertained by looking essentially at empirical, quantitative differences in the real world. This basic idea is that in interfirm comparisons, some way has to be found of meaningfully separating the innovative variety of a more behavioral (strategic) nature, from more "structural" (permanent) differences due to different markets and selection contexts in which firms operate. Empirical evidence is provided in order to show that not all observable diversity in performances reflect behavioral differences. Often such differences reflect differences in some key dimensions related to the i) industries, ii) technologies and iii) environmental conditions which define different contexts of competition between firms. The main implication for empirical research is that in the analysis of innovation-performance relationships in inter- firm comparisons the notion of variety is useful only to the extent that there are defined criteria for the selection of the firms. This in turn requires the previous identification of: i) the key dimensions defining the boundaries to firm strategies, and consequently ii) the level of aggregation at which at inter-firm similarities and differences of more strategic-behavioral nature can emerge.